UNIVERSITY of ROME "TOR VERGATA"

Transcription

1 UNIVERSITY of ROME "TOR VERGATA" Faculty of Engineering Ph. D. in MATERIALS FOR ENVIRONMENT AND ENERGY (XXI Cycle) Hybrid Polymer Electrolytes for Proton Exchange Membrane Fuel Cells: Synthesis and Applications Catia de Bonis A.A. 2008/2009 Advisor: Prof. Silvia Licoccia Coordinator: Prof. Enrico Traversa

2 Acknowledgements I would like to thank: Prof. Silvia Licoccia for her guidance, the insightful discussions and for the great possibility she offered me. This experience was really important for my professional and personal training. Prof. Enrico Traversa for giving me this opportunity and for freedom in conducting research. Dr. Alessandra D Epifanio and Dr. Barbara Mecheri for their assistance and helpful advice. Mrs. Cadia D Ottavi for her valuable technical assistance. Dr. M. Luisa Di Vona for her collaboration to this work. Prof. Marcella Trombetta of the University Campus Bio-Medico of Rome, for her contribution to this work. All members of the Chemistry and Engineering group of Tor Vergata University: Fang, Alma, Antonio, Laure, Chiara, Zakarya, Sherif, Alessandro, Nella, Emanuela, Andrea, Michele, Elisabetta, Debora, Vincenzo, Francesco, Corrado, Emiliana, Alessia, Simone, Milan, Alberto, Stefano and Marco, for their support and the enjoyable time we spent together. My family and Emanuele for their patience, encouragement, and support. This work has been financed by MIUR (FISR 2001 NUME Project Development of composite protonic membranes and innovative electrode configurations for Polymer Electrolyte Fuel Cells).

3 Abstract Proton exchange membrane fuel cells (PEMFCs) are promising power sources emerging among alternative energy conversion systems, because they can operate at relatively low temperature and offer numerous benefits, such as high efficiency, high power density and low polluting emissions. The present dissertation deals with the development of new proton conducting membranes having good conductivity, chemical and thermal stability, low methanol permeability and low cost. The main strategy used in this work was the preparation of sulfonated and silylated polyetheretherketone (PEEK) and polyphenylsulfone (PPSU) as membrane materials, because this synthetic approach represents a powerful tool to modulate the proton conductivity and hydrolytic stability of the electrolyte by the dosage of sulfonic acid groups and inorganic moieties covalently bound to the aromatic chains. Several types of proton exchange membranes were studied. Sulfonated and silylated PEEK and/or PPSU were used to prepare systems where two components resulted crosslinked by physical interactions or covalent bonds, obtaining the synergic effect of polymers having different conductivity and mechanical properties. Sulfonated and silylated polyetheretherketone PhSi 0.1 S 0.9 PEEK (degree of sulfonation DS=0.9, and degree of silylation DSi=0.1) was synthesized via (i) sulfonation of PEEK, (ii) conversion of sulfonated polyetheretherketone (S 0.9 PEEK) into sulfonyl chlorinated derivative (PEEKSO2Cl), (iii) lithiation of PEEKSO2Cl and subsequent addition of PhSiCl 3, followed by hydrolysis. The solubility of PEEKSO2Cl in organic solvent allows the silylation reaction to be carried out in homogeneous conditions. The structural characterization of the products by 1 H and 13 C NMR and ATR/FTIR spectroscopies highlighted the success of the synthetic pathway. The thermogravimetric analysis of PEEK derivatives indicated that the presence of the inorganic moieties stabilizes the aromatic matrix of the sulfonated polyetheretherketone. Blends of PhSi 0.1 S 0.9 PEEK and S 0.5 PEEK (DS=0.5) were prepared using different weight ratios of the two polymers. The membranes were characterized by water uptake measurements and electrochemical impedance spectroscopy (EIS). The results converge I

4 to indicate that the developed materials are promising electrolytes for PEMFC application. Silylated and sulfonated polyphenylsulfone PhSi 0.2 S 2 PPSU (DS=2.0 and DSi=0.2) was synthesized via (i) lithiation of PPSU and subsequent addition of PhSiCl 3, followed by hydrolysis, (ii) sulfonation by reaction with concentrated sulphuric acid. The chemical structure of polymers was investigated by 1 H and 13 C NMR, and ATR/FTIR, verifying the success of the developed synthetic route. Blends of PhSi 0.2 S 2 PPSU and S 0.5 PEEK were prepared, obtaining electrolytes with higher hydrolytic stability and increased proton conductivity with respect to those of pure S 0.5 PEEK membrane. Blend membranes showed also better performance in DMFC, where a reduced methanol permeability and adequately high power density values were observed, at temperature values as high as 100 C. All these features identify the prepared blend membranes as promising electrolytes for DMFC operating at intermediate temperatures. Two silylated and sulfonated PPSU derivatives: Si 0.2 S 2 PPSU (DS=2.0 and DSi=0.2) and Si 0.03 S 0.05 PPSU (DS=0.05 and DSi=0.03) were synthesized following two different routes. In the first one, PPSU was silylated by reaction with SiCl 4, then sulfonated by reaction with concentrated sulphuric acid, and Si 0.2 S 2 PPSU was obtained. In the second route, the use of the mild sulfonating agent ClSO 3 Si(CH3) 3 allowed a careful control of the degree of sulfonation, and PPSU with a lower DS was obtained. Subsequent silylation by reaction with SiCl 4 led to the final product Si 0.03 S 0.05 PPSU. An organic-inorganic hybrid polymer HSiSPPSU was synthesized by non-hydrolytic sol gel reaction of Si 0.2 S 2 PPSU and Si 0.03 S 0.05 PPSU. The condensation between the silanol groups of the two polymers led to the formation of Si-O-Si bonds, as highlighted by analysis of ATR/FTIR spectra. The electrochemical characterization of HSiSPPSU membranes by EIS showed adequately high conductivity values to make the hybrid polymer a suitable candidate for application in PEMFCs operating at T > 100 C. The strategies followed in this work seems to be an effective way to overcome some drawbacks related to conventional polymer membranes currently used, demonstrating the relevant role played by synthesis in the preparation of electrolytes for PEMFCs. II

5 Table of Contents 1 Introduction Power Generation from an Environmental Sustainability Perspective Research Objective 3 2 Proton Exchange Membrane Fuel Cells Fuel Cell Classification Polymer Electrolyte Membrane Fuel Cells Alkaline Fuel Cells Phosphoric Acid Fuel Cells Molten Carbonate Fuel Cells Solid Oxide Fuel Cells Operating Principle of PEMFC H 2 -PEMFC DMFC Performance Ideal Performance Actual Performance Efficiency PEMFC System Components Water Management Heat Management Methanol Crossover References 23 3 Proton Exchange Membranes Polymer Membrane Properties Tuneable Parameters Structural Features Transport Properties Nafion and Sulfonated Aromatic Polymers Membrane Structure as function of Water Content Proton Transport Proton Conduction Mechanisms Proton Conductivity and Water Diffusion Activation Energy Synthetic Approaches Composite and Blend Approaches References 51 4 Experimental Materials and Methods Starting Materials 56 III

6 4.1.2 Methods References PhSi 0.1 S 0.9 PEEK / S 0.5 PEEK Blend Polymers Synthesis Results and Discussion References PhSi 0.2 S 2 PPSU / S 0.5 PEEK Blend Polymers Synthesis Results and Discussion References PPSU-based Hybrid Electrolyte Polymers Synthesis Results and Discussion References Conclusions 107 List of Acronyms 110 List of Schemes 111 List of Tables 111 List of Figures 111 List of Papers 114 Riassunto 115 IV

7 1 Introduction 1.1 Power Generation from an Environmental Sustainability Perspective Throughout the world, the need of the hour is power generation coupled with environmental protection because energy use from traditional sources, such as burning fossil fuels, poses serious challenges to human health, energy security and sustainability of natural resources. Since the last century, alternative energy sources economy has been proposed as alternative to the current fossil-fuel economy. This alternative economy is based on fuel cells and other renewable sources like photovoltaic, hydro electric and wind power. Fuel cells are one of the most important devices because they convert the chemical energy of the fuel into electrical energy with high efficiency, minimal environmental pollution and low maintenance costs. They are one of the oldest electrical energy conversion technologies known to man because fuel cell experiments were performed by Sir William Grove as early as in Unfortunately, fuel cells were forgotten until the middle of the 20th century due to the invention of the combustion engine. They exhibited some renaissances in the 1960 s, in the US Apollo space program, and during the first oil crisis in 1973, when especially the interest for large power plants based on high-temperature fuel cells increased. In the last decades, the focus on environmental and pollution problems has made the world search for cleaner energy technology, and fuel cells have experienced increasing interest from governments and industry. Among the different type of fuel cells, Proton Exchange Membrane Fuel Cells (PEMFCs), offer great promise as a clean and efficient technology for energy generation and provide significant environmental benefits. Since PEMFCs are modular and have a simple design, they can be scaled up in size to suit the demands of a variety of applications. They have the potential to revolutionize transportation, which consumes 60% of the worldwide production of oil, improve stationary power generation, thereby reducing dependence on the electrical grid, and provide reliable portable power for our ever-increasing array of personal electronic devices. 1

8 1 - Introduction Conventional proton exchange membranes (PEM) are based on polymers in their protonated form. These materials are typically phase separated into a percolating network of hydrophilic nano-pores embedded in a hydrophobic polymer-rich phase domain. The hydrophilic nano-pores contain the acidic moieties which ensure the proton conductivity. The hydrophobic phase domain provides mechanical strength by stabilizing the morphology of the membrane. Nowadays, Nafion is the most widely used polymer electrolyte for PEMFCs due to its excellent oxidative stability, high proton conductivity and good mechanical strength within the operating environment of the cell. However, Nafion and similar perfluorinated polymers have relevant drawbacks, including decrease of conductivity at temperature over 80 C, high methanol permeability, and high cost that is incommensurate with potential mass markets. Consequently, a major trend in the present research activities is the preparation of alternative membrane materials which performance would be less affected by humidification, temperature and fuel crossover to overcome the barriers in the wide commercialization of PEMFCs. Among the different materials alternative to Nafion that are investigated, aromatic polymers are considered to have the proper features as membrane materials. They are low-cost polymers with good oxidative and thermal resistance and their chemical structure opens up a wide window of chemical modification opportunities, including the introduction of sulfonic moieties to obtain proton-conducting materials. In fact, the synthesis of functionalized polymers represents a very promising way to obtain electrolytes having the desired features because it allows to prepare tailored materials for specific applications. 2

9 1 - Introduction 1.2 Research Objective The purpose of the present dissertation is to develop new polymer electrolytes with good proton conductivity and the required chemical and morphological stability for PEMFCs applications. The aromatic polymers Polyetheretherketone (PEEK) and Polyphenylsulfone (PPSU) were chosen as starting materials due to their promising features as membrane materials. These polymers show excellent chemical resistance, high thermo-oxidative stability, good mechanical properties, ease of direct functionalization and low cost. However, in these systems the high degree of sulfonation required to attain proper conductivity values leads to highly swollen or even water soluble products, lowering mechanical and morphological properties. Hence, several approaches were followed to modulate the ratio between hydrophilic and hydrophobic domains: such ratio, in fact, determines hydrolytic stability and conductivity, which balance will, eventually, be responsible of the electrolyte performance under fuel cell operating conditions. Different systems were thus studied: Derivatives of PEEK and PPSU with silicon-containing moieties covalently linked to the aromatic chain were synthesized. The used synthetic strategies allowed to tailor the molecular structures controlling the amount of inorganic moieties to obtain polymers with high sulfonation degree and hydrolytic stability. Blend membranes where the microstructure of sulfonated PEEK, the major component, was modified by physical cross-linking with silylated and sulfonated PPSU or PEEK. Polymer systems where two derivatives of PPSU having different degrees of sulfonation and silylation were cross-linked through Si-O-Si moieties. In all cases, good membrane performance were obtained due to the interactions between the different components that avoided the degradation of highly sulfonated polymers, maintaining good proton conductivity. 3

10 2 Proton Exchange Membrane Fuel Cells 2.1 Fuel Cell Classification A Fuel Cell is an electrochemical device similar to a battery; it transforms the chemical energy in Direct electric Current (DC) which can be used to power vehicles or to run household appliances. The main difference with respect to a conventional battery is that in the battery the charge is consumed during the discharge and restored with the recharges. The fuel cell functions continuously until there is the fuel (reactants on the electrodes) as a standard combustion motor, because the power is assured by the chemical reaction between the reactants (fuel and oxidant). Fuel cells are classified primarily by the kind of used electrolyte. This determines the chemical reactions that take place in the cell, the used fuel, the kind of catalysts required, the temperature range in which the cell operates, and other factors. A second grouping can be done considering the operating temperature of the fuel cells. Low-temperature fuel cells operate below about 250 C and include Proton Exchange Membrane Fuel Cells, Alkaline Fuel Cells and Phosphoric Acid Fuel Cells. High-temperature fuel cells operate in the range C and include Solid Oxide Fuel Cells and Molten Carbonate Fuel Cells. 1 These characteristics affect the applications for which these cells are most suitable. Each fuel cell currently under development has its own advantages, limitations and potential applications. An overview of the fuel cell types is given in Table

11 2-Proton Exchange Membrane Fuel Cells Table Fuel Cell Classification Polymer Electrolyte Membrane Fuel Cells Proton exchange membrane fuel cells (PEMFCs) use a solid polymer as electrolyte and porous carbon electrodes containing a platinum catalyst. Low operating temperature allows PEMFCs to start quickly and results in low wear on system components. Depending on the fuel, a PEMFC can be classified as direct H 2 -PEMFC that is directly fed with hydrogen, indirect reformate PEMFC where the hydrogen fuel is produced by steam reforming, and direct methanol fuel cell (DMFC) that uses methanol as fuel. H 2 -PEMFC H 2 -PEMFCs are used to power automobiles, scooters, buses, backup power generators, boats and underwater vehicles. 5

12 2-Proton Exchange Membrane Fuel Cells This kind of PEMFC uses hydrogen as fuel and oxygen as an oxidant, giving only water as waste product, but whereas oxygen can be taken from the surrounding air, the availability of hydrogen presents a challenge. Most fuel cell vehicles powered by pure hydrogen must store the hydrogen onboard as a compressed gas in pressurized tanks. Due to the low energy density of hydrogen, it is difficult to store enough hydrogen onboard to allow vehicles to travel the same distance as gasoline-powered vehicles before refuelling, typically miles. 2 Alternatively, cryogenic hydrogen (liquid) can be stored; this process would reduce the low-density problem, but the liquefaction uses about 30% of the inherent energy. 3 Reformate Fuel Cell The problem of storage and transportation of hydrogen can be solved by producing it from a steam reforming process or by a partial oxidation of liquid fuels such as methanol, ethanol, by using a fuel processor placed onboard the fuel cell powered vehicle. Basically, all organic molecules are potential candidates for a direct electrochemical conversion. However, due to undesired side reactions and high activation overpotentials, the rate of electrochemical conversion is often very poor. The steam reforming reactions for the methanol are shown below: Methanol Steam Reforming: CH 3 OH (g) + H 2 O CO 2(g) + 3 H 2(g) (Eq. 2.1) Water Gas Shift Reaction: CO + H 2 O CO 2 + H 2 (Eq. 2.2) Selective Oxidation of CO: CO + ½ O 2 CO 2 (Eq. 2.3) The main problem connected to use of a reformed fuel is the platinum catalyst poisoning by CO, which causes a decrease of the electrocatalytic activity of the noble metal. Moreover, the reformer also releases carbon dioxide (a greenhouse gas), though less than that emitted from current gasoline-powered engines. 4 6

13 2-Proton Exchange Membrane Fuel Cells DMFC In direct methanol fuel cells the liquid fuel is directly oxidized at the anode, so that these devices have the advantage of being simpler power systems in comparison to fuel cells using H 2 derived from methanol reforming. Compared with an H 2 -PEMFC, a DMFC shows a larger efficiency decrease due to greater polarization losses and high methanol permeability of the polymer electrolyte (up to 50% methanol crossover). 5 DMFCs have potential application in portable devices, such as laptops, mobile phones and MP3 players that require low current density and small system size Alkaline Fuel Cells Alkaline fuel cells (AFCs) were among the first fuel cell technologies developed, and the first type widely used in the USA space program to produce electrical energy and water onboard spacecrafts. AFCs use a solution of potassium hydroxide in water as the electrolyte and can use a variety of non-precious metals as catalysts at the anode and cathode. The disadvantage of this fuel cell type is that it is easily poisoned by carbon dioxide. In fact, even the small amount of CO 2 in the air can affect the operation and lifetime of the cell, adding to cost Phosphoric Acid Fuel Cells Phosphoric acid fuel cells (PAFCs) are typically used for stationary power generation, but some of them have been used to power large vehicles such as city buses. This type of fuel cells use liquid phosphoric acid as electrolyte and porous carbon electrodes containing a platinum catalyst. PAFCs are more tolerant of impurities in fossil fuels that have been reformed into hydrogen than PEMFCs, but are less powerful than other fuel cells, given the same weight and volume. As a result, these fuel cells are typically large and heavy Molten Carbonate Fuel Cells Molten carbonate fuel cells (MCFCs) are currently being developed for industrial and military applications. 7

14 2-Proton Exchange Membrane Fuel Cells They are high-temperature fuel cells that use an electrolyte composed of a molten carbonate salt mixture suspended in a porous, chemically inert ceramic lithium aluminium oxide matrix. Since they operate at extremely high temperatures, nonprecious metals can be used as catalysts at the anode and cathode, reducing costs. The primary disadvantage of current MCFC technology is durability. The high temperatures at which these cells operate and the corrosive electrolyte used accelerate component breakdown and corrosion, decreasing cell life Solid Oxide Fuel Cells Solid oxide fuel cells (SOFCs) are suitable for stationary applications as well as for auxiliary power units used in vehicles to power electronics. They use a hard, non-porous ceramic compound as electrolyte. High operating temperature removes the need for precious-metal catalyst and allows SOFCs to reform fuels internally, reducing cost. However, high-temperature operation results in a slow start-up and requires significant thermal shielding to retain heat and protect personnel. 6 8

15 2-Proton Exchange Membrane Fuel Cells 2.2 Operating Principle of PEMFC H 2 -PEMFC Proton exchange membrane fuel cell consists of two porous carbon electrodes, the anode and the cathode, separated by a polymer electrolyte, the proton conducting membrane. The electrodes are connected to an external load circuit and integrated between each electrode and the membrane is a thin layer of platinum catalyst. A schematic representation of H 2 -PEMFC is shown in Figure Figure Schematic diagram of H 2 -PEMFC. On the anode side, hydrogen diffuses to the anode catalyst where it dissociates into protons and electrons. The protons are conducted through the membrane to the cathode, while the electrons are forced to travel in an external supplying power circuit because the membrane is electrically insulating. On the cathode catalyst, oxygen molecules react with the electrons and protons to form water, the only waste product. 7 Hydrogen Oxidation The overall reaction of hydrogen oxidation is: H 2 2 H e - (Eq. 2.4) 9

16 2-Proton Exchange Membrane Fuel Cells The mechanism of H 2 electro-oxidation on Pt in acid electrolytes is thought to proceed by the absorption of the gas onto the catalyst surface, followed by charge transfer which gives hydrogen ions, as follows: H Pt 2 Pt-H ads (Eq. 2.5) 2 Pt-H ads 2 Pt + 2 H e - (Eq. 2.6) When pure hydrogen is used as fuel the performance of the anode is excellent with pure Pt. Unfortunately in most practical systems the fuel stream contains traces of elements or compounds, such as CO, S, and NH 3, that can poison the anode catalysts. In particular, CO is one of the major poisons in low temperature fuel cells, due to its absorption to the active sites of the platinum based catalysts so that no, or virtually no, sites are available for reaction with H 2. 8 Oxygen Reduction reduction: Each oxygen molecule requires the transfer of four electrons for complete O H e - 2 H 2 O (Eq. 2.7) The simultaneous transfer of these electrons is highly unlikely, in fact, partial electron transfer takes place leading to the formation of intermediates such as superoxide. The application of a platinum electrocatalyst allows the stabilization of these intermediates and the reaction then proceeds at a reasonable and useful rate. The mechanism is believed to take place through the following sequence: 5 O 2 + Pt Pt-O 2 (Eq. 2.8) Pt-O 2 + H + + e - Pt-HO 2 (Eq. 2.9) Pt-HO 2 + Pt Pt-OH + Pt-O (Eq. 2.10) Pt-OH + Pt-O + 3 H e - 2 Pt + 2 H 2 O (Eq. 2.11) 10

17 2-Proton Exchange Membrane Fuel Cells Although a huge variety of catalysts have been investigated, at present, Pt based catalysts are the most widely used cathode materials in low temperature fuel cells, due to their intrinsic activity and stability in acidic solutions DMFC In DMFC, methanol and water electrochemically react at the anode to produce carbon dioxide, proton and electrons. The protons are conducted through the membrane to the cathode, while the electrons travel through the external circuit where they can be made to do useful work, such as powering an electric motor. 9 On the cathode catalyst, oxygen molecules react with the electrons and protons to form water, analogously to that which occurs in a H 2 -PEMFC (Eq. 2.7). Methanol Oxidation electrons: The electro-oxidation of methanol to carbon dioxide involves the transfer of six CH 3 OH + H 2 O CO H e - (Eq. 2.12) Few materials are able to electrosorb methanol, and in acidic electrolytes, only platinum-based electrocatalysts have shown the required activity and chemical stability. 10 The adsorption mechanism is believed to take place through the sequence of step shown in Figure 2.2. Figure Scheme of methanol electrosorption mechanism on pure Pt surface. 11

18 2-Proton Exchange Membrane Fuel Cells The mechanism shows the electrosorption of methanol on the surface of platinum with sequential proton and electron stripping, leading to the main catalyst poison, linearly bonded carbon monoxide (Pt-CO). Subsequent reactions are believed to involve oxygen transfer to the Pt-CO species to produces CO Much research is being rendered on catalysts for methanol oxidation to find a catalyst which can avoid the poisoning effect of the CO species. Several promoters, such as Ru, Sn, W, have been found to increase the activity of the Pt catalyst. At the present, the most active methanol electro-oxidation catalysts are bimetallic alloys consisting of Pt and Ru supported on carbon Performance Ideal Performance A fuel cell is a galvanic cell which converts the chemical energy of the reaction directly into the electrical energy. The maximum electrical work (W el ) obtainable in a fuel cell operating at constant temperature and pressure is given by the change in Gibbs free energy (ΔG) of the electrochemical reaction: W el = ΔG = - nfe (Eq. 2.13) where n is the number of electrons involved in the reaction, F is the Faraday s constant and E is the ideal potential. The thermodynamic equilibrium potential E represents the ideal performance of the cell and is described by the Nernst equation: RT Π[reactant activity] E = E + ln nf Π[product activity] (Eq. 2.14) where E is the ideal potential at standard conditions ( K and 1 atm), resulting from the difference between the potentials of cathode and anode. For H 2 -PEMFC, the standard ideal potential is 1.23 volts (V) with liquid water as product and 1.18 V with gaseous water as product: 13 12

19 2-Proton Exchange Membrane Fuel Cells 2 H 2 4 H e - (E anode = 0.00 V) (Eq. 2.15) O H e - 2 H 2 O (l) (E cathode = 1.23 V) (Eq. 2.16) 2 H 2 + O 2 2 H 2 O (l) (E = 1.23 V) (Eq. 2.17) For DMFC, the standard ideal potential corresponds to 1.18 volts: CH 3 OH + H 2 O CO H e - (E anode = 0.05 V) (Eq. 2.18) 3/2 O H e - 3 H 2 O (l) (E cathode = 1.23 V) (Eq. 2.19) CH 3 OH + 3/2 O 2 CO H 2 O (l) (E = 1.18 V) (Eq. 2.20) Actual Performance When current flows in the cell, the actual potential (V) is decreased from its ideal value (E) because of irreversible losses, called polarization or overpotential (η): V = E - η (Eq. 2.21) These losses originate primarily from three sources: (1) activation polarization (η act ), related to the rates of electrochemical reactions; (2) ohmic polarization (η ohm ), due to resistance of the cell components and interconnects; (3) concentration polarization (η conc ), arising from the depletion of reactants at catalyst sites under high loads. As can be seen in Figure 2.3, the activation polarization loss (η act ) is dominant at low current density, giving the characteristic logarithmic shape to the polarization curve. It is present when the rate of an electrochemical reaction at an electrode surface is controlled by sluggish electrode kinetics. In other words, activation polarization is directly related to the rates of electrochemical reactions and therefore to the activation barriers that must be overcome by the reacting species prior to current and ion flow. Ohmic losses (η ohm ) occur because of resistance to the flow of ions in the electrolyte and resistance to flow of electrons through the electrode materials. Because both the electrolyte and electrodes obey Ohm's law, the ohmic losses can be expressed by the equation: 13

20 2-Proton Exchange Membrane Fuel Cells η ohm = i R (Eq. 2.22) where i is the current flowing through the cell, and R is the overall resistance of the cell (protonic, electronic and contact resistance). Ohmic polarization varies directly with current, increasing over the whole range of current because cell resistance remains essentially constant. The ohmic losses through the electrolyte can be reduced by decreasing the electrode separation, that is by reducing membrane thickness, and enhancing the ionic conductivity of the electrolyte. The concentration losses (η conc ) occur over the entire range of current density, but these losses become prominent at high limiting currents where it becomes difficult to provide enough reactant flow to the cell reaction sites. The slow transport of reactants and products to and from the electrochemical reaction sites causes a loss of potential due to the inability of the surrounding material to maintain the initial concentration of the bulk fluid. Figure Ideal and Actual fuel cell Voltage/Current density characteristic. 14

21 2-Proton Exchange Membrane Fuel Cells Moreover, when no current is drawn from the cell, the actual cell potential, i.e. open circuit voltage (OCV), does not reach the thermodynamic potential value E because of parasitic electrochemical processes. Some voltage loss at the OCV is due to crossover of some hydrogen or methanol through the membrane to the cathode where some hydrogen or methanol oxidation competes with oxygen reduction, causing a mixed potential Efficiency The thermal efficiency (ε thermal ) of an energy conversion device is defined as the amount of useful energy produced relative to the change in stored chemical energy (commonly referred to as thermal energy) that is released when a fuel is reacted with an oxidant: useful energy ε thermal = H (Eq. 2.23) In order to compare fuel cells with other systems, such as internal combustion engines, an evaluation of the system efficiency is necessary. In the internal combustion engines the conversion of chemical energy into electricity is not a direct process, since chemical energy is firstly converted into mechanical energy, and then transformed into electrical energy: Chemical energy Heat Mechanical energy Electrical energy The combustion of a hydrocarbon (chemical energy) is accompanied by a rise in temperature as these reactions are exothermic and the reaction products are usually gases. The heat increase causes an expansion of the formed gases, which in turn can produce mechanical work that be transformed into electrical energy by means of a rotating generator. The maximum thermal efficiency for such conversion systems is determined by Carnot s law and it is written as follows: 15

22 2-Proton Exchange Membrane Fuel Cells ε thermal Wr T = = 1 H T 2 1 (Eq. 2.24) where W r is the reversible work performed, ΔH is the enthalpy change of reaction, T 1 is the higher temperature of the heat source and T 2 is the lower temperature, e.g. at the gas outlet. In general, these efficiencies are not surpassing 40 % for the most efficient engines. Fuel cells represent an attractive alternative as their efficiency is not limited by the Carnot cycle and most of the chemical energy stored in the reactants can be converted into electricity: Chemical energy Electrical energy For a fuel cell, the ideal efficiency (ε ideal ) can be calculated from the Gibbs free energy change (ΔG) and the enthalpy change (ΔH) of the electrochemical reaction. It describes the ratio between the maximum electric work that can be drawn from a fuel cell (W el ), and the overall quantity of energy that can be transformed into heat (ΔH). Ideally, the free energy of the reaction can be completely converted into electrical energy and the efficiency is given by: -Wel ΔG ε ideal = = H ΔH (Eq. 2.25) Thus, the efficiency of an ideal fuel cell operating reversibly on pure hydrogen and oxygen at standard conditions (1 atm pressure and 25 ºC) would be The efficiency of an actual fuel cell (ε actual ), can be expressed in terms of the ratio of the operating cell voltage (V) to the ideal cell voltage (E), according to the equation report below: ε actual 0.83 V = (Eq. 2.26) E 16

23 2-Proton Exchange Membrane Fuel Cells where the actual cell voltage is less than the ideal cell voltage because of the losses due to electrode overpotentials and the electrolyte resistance. A fuel cell can be operated at different current densities and the corresponding cell voltage then determines the actual efficiency. A typical H 2 -PEMFC running at 0.7 V has an actual efficiency of about 0.5, meaning that 50% of the energy content of the hydrogen is converted into electrical energy and the remaining 50% will be converted into heat. 17

24 2-Proton Exchange Membrane Fuel Cells 2.4 PEMFC System Components The electrodes and membrane together form the membrane electrode assembly (MEA). A single cell is realized when bipolar flow field plates, with gas channels grated into their surface, are placed on each side of the MEA, as shown in Figure 2.4. Figure Scheme of a single PEMFC. The anode and cathode are thin sheets of porous carbon paper or carbon cloth having a catalyst layer pressed onto one side of them (typically Pt or Pt/Ru alloys) and wetproofed with Teflon. The side containing the catalyst is placed in contact with the proton exchange membrane. 18

25 2-Proton Exchange Membrane Fuel Cells Polymer Electrolyte Solid polymer electrolytes form a thin but sound electronic insulator and gas barrier between the two electrodes while allowing rapid proton transport and high current densities. The reduction of the membrane thickness can lead to the greatest improvement in fuel cell performance. The advantages gained with this simple strategy include lower membrane resistance, lower material utilization, and improved hydration of the entire membrane. However, the extent to which a membrane can be thinned is limited because of difficulties with durability and reactant crossover. This is especially true for direct methanol fuel cells, where excessive methanol transport occurs, reducing power density. 14 The essential requirements of polymer electrolyte membranes include: good proton conductivity, chemical and thermal stability, low methanol crossover (in DMFCs), and low cost. 15 A detailed discussion on electrolyte features and membrane materials has been developed in Chapter 3. Electrodes Electrodes for PEMFCs are generally porous gas diffusion electrodes to ensure the supply of the reactant gases to the active zones where the noble metal catalyst is in contact with the ionic and electronic conductor. The main requirement of a good electrode is a three-phase boundary between the gas supply on the one hand, and the catalyst particle and the ionic conductor, on the other hand. Figure 2.5 shows a detailed scheme of a typical electrode, which displays three layers: (i) teflonized substrate (typically carbon cloth); (ii) a diffusion layer, generally formed by carbon particles of about 0.1 µm size along with Teflon and (iii) an active layer, where Pt catalyst grains (dimensions 2-4 nm) are supported on carbon particles (Pt load usually 0.4 mg/cm 2 or less) with or without Teflon; a proton conducting membrane phase partly penetrates the pores in this layer

26 2-Proton Exchange Membrane Fuel Cells Figure Scheme of a typical electrode. The catalyst particles must be in direct contact with an electronic conductor to ensure that the electrons are supplied to or taken away from the reaction site. Electronic conductivity is usually provided by carbon particles. To enhance the three-phase boundary and to obtain a maximum utilization of the highly dispersed electrocatalyst, additional ionomer (usually liquid Nafion) is impregnated into the active layer. This ensures a wide access of the H + ions within the catalyst layer. When using humidified gases or a methanol/water mixture (for DMFC) the catalyst layer must be sufficiently hydrophobic to prevent the pores from flooding. This hydrophobicity can be provided by introducing polytetrafluoroethylene (PTFE) as a binder, in combination with Nafion that is hydrophilic. 17 Additional Components Bipolar plates supply the reactants through the flow channels to the electrodes and serve the purpose of electronically connecting one cell to another in the electrochemical cell stack. Their essential requirements are: high values of electronic conductivity, high mechanical strength, impermeability to reactants, resistance to corrosion and low cost of automated production

27 2-Proton Exchange Membrane Fuel Cells Generally, bipolar plates are made up of graphite and gas channels are grated into their surface in a parallel flow configuration. Both the material of the flow field plates and the design of the flow field are subject to several development projects considering volume, weight, pressure loss etc Water Management Water is carried into the fuel cell via humidified gas streams and methanol/water mixture (for DMFCs) entering gas diffusion electrodes. An additional source of water is the oxygen reduction reaction occurring at the cathode. Water transport across the membrane takes place by three different processes: electro-osmotic drag, diffusion and hydraulic permeation, which are schematically depicted in Figure Figure Scheme illustrating the modes of water transport in an operating PEMFC. The electro-osmotic drag is due protons that migrate across the membrane from the anode to cathode carrying water molecules with them. The production of water by the oxygen reduction reaction together with the electroosmotic drag creates an excess of water at the cathode. This gradient of water content across the membrane may results in back diffusion of water from cathode to anode. 21 If a differential pressure exists across the membrane, a hydraulic permeation can also contribute to the water transport. 21

28 2-Proton Exchange Membrane Fuel Cells In the absence of the latter, the net water flux across the membrane is a combination of diffusion and electro-osmotic drag and two specific water transport parameters have then a direct impact on cell performance: the water self-diffusion coefficient (DH 2O) that is a measure of the average mobility of water in the membrane, and the electro-osmotic drag coefficient (K drag ) that represents the average number of water molecules that are carried through the membrane per proton conducted. In PEMFC, the membrane to be well humidified because the proton conducting process relies on water content of membrane. Operating at high current density, the anode side must be humidified by pre-humidifying the hydrogen to avoid dehydration due to the electro-osmotic drag. Moreover, the cathode side needs to remove water to prevent slow oxygen diffusion due to flooding of the catalytic layer. Water removal conditions should be carefully controlled. In fact, if water is removed too quickly, the membrane dries and resistance across it increases, whilst, if the water is removed too slowly, the cathode will flood preventing the oxygen from reaching the catalyst and stopping the reaction Heat Management Increase in operating temperature of fuel cell is attractive for a number of reasons: (i) improved tolerance of the electrodes to carbon monoxide, which enables the use of hydrogen produced by reforming of natural gas, methanol or gasoline; (ii) simplification of the cooling system; (iii) increased proton conductivity; and (iv) in DMFC, improved kinetics of the methanol oxidation reaction at the anode. 22 However, increase in temperature confers the restraints. Because in the most of proton exchange polymer materials the conduction occurs through a water-assisted mechanism, they exhibit the highest proton conductivity when fully hydrated. This requires humidification of gas feeds before entering the fuel cell and the application of pressure to maintain adequate relative humidity Methanol Crossover The methanol crossover consists of the protonic drag of methanol through the membrane of the DMFC, similar to electro-osmotic drag of water. In fact, due to the similar properties of water, such as dipole moment, methanol molecules are transported to the cathode by the electro-osmotic drag as well as diffusion. 22

29 2-Proton Exchange Membrane Fuel Cells The effects of methanol crossover on DMFC performance consist of (i) limited activity of the anode catalyst, (ii) lowering of methanol utilization, resulting of methanol/oxygen recombination at the cathode catalyst, and (iii) the deleterious effects of methanol penetrating the cathode on the performance of the cathode itself. 5 Methanol crossover depends on several factors: the membrane permeability, the concentration of methanol in the fuel feed, the operating temperature and the performance of the anode itself. The nature of the membrane is a very important factor and great efforts have been made to vary the membranes morphological features in order to reduce their methanol permeability. 23 The crossover is also dependent on the methanol concentration in the feed. A higher concentration as well as a higher temperature in the cell increase the diffusion of methanol through the membrane, lowering the cell performance. This factor can be controlled to a certain extent by strictly correlating methanol feed concentration with the actual current demand of the cell. An optimum concentration of methanol was considered to be around 1-2 M methanol in water. 24 Finally, if the anode is sufficiently active to oxidize methanol at a rate comparable or higher than the rate of methanol supply, the concentration at interface could decrease to zero preventing methanol to cross through the membrane. The anode activity can be promoted by the use of suitable Pt/Ru catalysts. 25 Slight activity improvements can be found by the use of multinary catalysts References 1) L. Carrette, K.A. Friedrich, U. Stimming, Fuel Cells 2001, 1, 5. 2) F. Barbir, PEM Fuel Cells: Theory and Practice, Elsevier Academic Press, New York, ) K. Kordesch, G. Simader, Fuel Cells and their Application, 1 st ed., VCH, Weinheim, ) R. Devanathan, Energy Environ. Sci. 2008, 1, ) A.S. Aricò, S. Srinivasan, V. Antonucci, Fuel Cells 2001, 1, ) C. Song, Catalysis Today 2002, 77,

30 2-Proton Exchange Membrane Fuel Cells 7) P.R. Pathapati, X. Xue, J. Tang, Renewable Energy 2005, 30, 1. 8) P. Futerko, I.M. Hsing, Electrochim. Acta 2000, 45, ) M. Hogarth, Fuel Cell Technology Handbook: Prospects of the Direct Methanol Fuel Cell, CRC Press LLC, ) R. Parson, T. Vandernoot, J. Electroanal. Chem. 1988, 257, 9. 11) P.A. Christensen, A. Hamnett, R.J. Potter, Ber. Bunsenges. Phys. Chem. 1990, 94, ) A.S. Aricò, V. Baglio, A. Di Blasi, E. Modica, P.L. Antonucci, V. Antonucci, J. Electroanal. Chem. 2003, 557, ) P.W. Atkins, Physical Chemistry, 3 rd ed., W.H. Freeman and Company, New York, NY, ) K. Prater, J. Power Sources 1990, 29, ) M. Hickner, H. Ghassemi, Y.S. Kim, B.R. Einsla, J.E. McGrath, Chem. Rev. 2004, 104, ) P. Costamagna, S. Srinivasan, J. Power Sources 2001, 102, ) R. Dillon, S. Srinivasan, A.S. Aricò, V. Antonucci, J. Power Sources 2004, 127, ) G. Hoogers, Fuel Cell Technology Handbook: Fuel Cell Components and Their Impact on Performance, CRC Press LLC, ) P.L. Hentall, J.B. Lakeman, G.O. Mepsted, P.L. Adcock, J.M. Moore, J. Power Sources 1999, 80, ) J. Zhang, Z. Xie, J. Zhang, Y. Tang, C. Song, T. Navessin, Z. Shi, D. Song, H. Wang, D.P. Wilkinson, Z.-S. Liu, S. Holdcroft, J. Power Source 2006, 160, ) T.A. Zawodzinski, C. Derouin, S. Radzinski, R.J. Sherman, V.T. Smith, T.E. Springer, S. Gottesfeld, J. Electrochem. Soc. 1993, 140, ) S. Cleghon, J. Kolde, W. Liu, Handbook of Fuel Cells-Fundamentals, Technology and Application, J. Wiley and Sons, Ltd, Chichester, ) V. Baglio, A. Di Blasi, A.S. Aricò, V. Antonucci, P.L. Antonucci, C. Traknprapai, V. Esposito, S. Licoccia, E. Traversa, J. Electrochem. Soc. 2005, 152, A ) X. Ren, T.E. Sprinter, S. Gottesfeld, J. Electrochem. Soc. 2000, 147, ) B.D. McNicol, D.A.J. Rand, K.R. Williams, J. Power Sources 1999, 83,

31 2-Proton Exchange Membrane Fuel Cells 26) A.S. Aricò, Z. Poltarzeski, H. Kim, A. Morana, N. Giordano, V. Antonucci, J. Power Sources 1995, 55,

32 3 Proton Exchange Membranes 3.1 Polymer Membrane Properties The membrane is the core component of PEMFCs. Its function is twofold: proton conduction from the anode to the cathode, and effective separation of the electrode reactants. 1 To achieve good cell performance, the membrane must possess the following features: 2 high proton conductivity, greater than 10-2 S cm -1 for a fully hydrated polymer; low permeability to reactants; chemical and electrochemical stability under operating conditions; mechanical properties (strength, flexibility and processability); production costs compatible with intended application. To focus on the relevant properties of polymer electrolytes, it is useful to categorize them in tuneable parameters, structural features and transport properties, as summarized in Table Tuneable Parameters Structural Features Transport Properties Backbone Characteristics Acid Group Characteristics Water Domain Size, Shape, Interface Interanionic Distance Proton Conductivity Water Transport Properties Ion Exchange Capacity Water Uptake Reactant Permeability Processing Tortuosity Table Relevant membrane properties. 26

33 3-Proton Exchange Membranes Tuneable Parameters The tuneable parameters represent the characteristics that can be controlled or modified in the preparation of the electrolyte. They include backbone and acid group characteristics, ion exchange capacity and processing conditions. Backbone characteristics offer perhaps the greatest number of controllable variables due to the extreme versatility of chemical synthetic processes. The nature of the polymer backbone is extremely important in determining its flexibility, strength and processability, oxidation and thermal resistance. 4 The acid group is another modifiable component of the polymer. Most polymer electrolytes have acid groups covalently bound to the polymer backbone, either through a side chain or directly through a chemical bond. In the case of side chains, features such as the length of the side chain, its flexibility, and its hydrophobicity are also important. 5 Ion-exchange capacity (IEC) indicates the density of ionizable hydrophilic groups in the membrane matrix, which are responsible for the ionic conductivity. IEC is probably the simplest quantity to control from a synthetic standpoint. 6 Processing of polymer electrolyte includes methods of preparation and treatment of membrane. Processing conditions affect polymer matrix morphology and hence characteristics of membrane such as water uptake, mechanical properties and proton conductivity Structural Features The structural features that determine the behaviour of water within the membrane are correlated to the water domain size, shape and the nature of the domain interface. Interanionic distances among the sulfonate sites are a contributing factor to water domain features and primarily affect the proton conductivity of the membrane. The energy barriers associated with proton transport are also dependent on the distance between anionic sites, so that the greater the distance between sites the greater the resistive losses associated with transport. 8 Tortuosity or continuity of the phases is important for both transport properties and mechanical stability. Tortuosity affects transport properties by describing the path 27

34 3-Proton Exchange Membranes length over which transport occurs and mechanical properties, because tortuosity also addresses the architecture of the structural phase Transport Properties Proton conductivity is fundamental for any fuel cell electrolyte and is usually the first characteristic considered when evaluating membranes for potential use. It is well know that for these systems proton conductivity is strongly dependent on water content, and performance improvements are needed in the area of increased conduction at low water content, particularly for higher temperature applications. 4 Cell performance depends also on two specific water transport parameters, the water self-diffusion coefficient (DH 2O) and the electro-osmotic drag coefficient (K drag ). The first coefficient is a measure of the average mobility of water in the membrane and the latter is a measure of the number of water molecules that are carried through the membrane per proton conducted. These coefficients are particularly important for water management both on a local and global scale. On a local scale, they affect hydration of the electrolyte and the electrodes which influences conductivity, electrode kinetics and mass transport, due to flooding. On a global scale, they affect feed gas flow rates, operating pressures, and humidification and cooling requirements. Reactant permeability is another important transport property in these materials. While hydrogen and oxygen permeability in polymer electrolyte membranes are typically low, methanol permeability represents a significant problem in DMFCs. In such fuel cells, methanol permeability leads to significant performance losses due to crossover, impacting the cathode kinetics and overall fuel efficiency. 9 28

35 3-Proton Exchange Membranes 3.2 Nafion and Sulfonated Aromatic Polymers Nafion was developed 40 years ago by DuPont Inc. 10 It consists of a polytetrafluoroethylene (PTFE) backbone, which gives it high chemical inertness. The side chains consist of perfluorinated vinyl polyether ending in sulfonic acid groups - SO 3 H that give proton exchange capability to the polymer. The chemical structure of Nafion is shown in Figure 3.1, where the values of n, x and y can be varied to produce materials with different equivalent weights. Figure Chemical structure of Nafion. Among non-fluorinated polymers alternative to Nafion, arylene main-chain polymers can provide some definitive advantages. Such polymers are inexpensive and open up a wide window of chemical modification opportunities. Their aromatic structure offers the possibility of electrophilic as well as nucleophilic substitution, allowing to obtain material with desired features for PEMFC applications. 11,12 In particular, the most important modification regards the introduction of sulfonic moieties to obtain proton-conducting materials. Commercial polymers from the aromatic family are often of the type poly-pphenylene shown in Figure 3.2a, where X is an atom or group of atoms, giving the polymer chains a certain degree of flexibility and hence processability

36 3-Proton Exchange Membranes Most simply, X is a single atom such as S in polyphenylene sulfide (PPS) (Figure 3.2b) or O in polyphenylene oxide (PPO) (Figure 3.2c). In particular, ether links provide a very good choice of functional groups as the -C-O-C- link itself is very flexible and also is highly resistant to thermal oxidation. In the development of polymer electrolyte membranes, modifications of the aromatic polymers containing ether links, such as polyetheretherketone (PEEK) (Figure 3.2d), have been widely investigated. 14,15 (a) Poly-p-phenylene (b) Polyphenylene sulfide (c) Polyphenylene oxide (d) Polyetheretherketone (e) Udel Polysulfone (f) Poly(2,2 -m-(phenylene)-5,5 -bibenzimidazole) Figure Polymer structures of interest. More commonly, X is a simple functional group such as SO 2 in polysulfone, NHCO in polyamides, COO in polyesters, and CO in polyketones. 30

37 3-Proton Exchange Membranes Modifications of polymers of the polysulfones (PSUs) family, such as Udel Polysulfone shown in Figure 3.2e, have been widely studied by several research groups. 16,17 The polybenzimidazoles represent another group of high performance aryl polymers. One of these is the poly(2,2 -m-(phenylene)-5,5 -bibenzimidazole) (PBI) whose structure is reported in Figure 3.2f. Non-modified PBI shows some tendency to take up water but its proton conductivity is very low. Hence, it is necessary to dope the polymer with sulphuric or phosphoric acid to increase its proton conductivity. 18,19 31

38 3-Proton Exchange Membranes 3.3 Membrane Structure as function of Water Content Modeling of water sorption and effect on the membrane structure is important since proton conductivity is strongly dependent on the water content. To examine the processes that take place, it is useful to focus on Nafion, for which data are more readily available. A comparison with sulfonated polyetherketones membranes is then reported. The general structure of Nafion membrane as a function of water content has been the topic of many studies, such as swelling, infrared, X-ray and microscopy studies, just to name a few. 20,21 These have shown that a hydrated membrane contains two phases, an ionic phase that is associated with the hydrated sulfonic acid groups, and a non-ionic phase that is the perfluorinated matrix. The actual form of the phases depends on the water content. Several models have been proposed since the early 1970s, to predict ionic transport properties of Nafion describing the way in which ionic groups aggregate. These models include the Mauritz-Hopfinger Model, 22 the Yeager Three Phase Model and the Gierke Cluster Network Model. The Yeager Model is based on a three-phase clustered system with interconnecting channels within the polymer. The three regions consist of (A) a fluorocarbon backbone, (B) an interfacial region and (C) the clustered regions, as shown in Figure 3.3. Figure Yeager s model of Nafion membranes: fluorocarbon region (A), interfacial zone (B), ionic cluster region (C). 32

39 3-Proton Exchange Membranes The hydrophobic region (A) contains the CF 2 -CF 2 -CF 2 units. The interface region (B) contains the side chains O-CF 2 -O-CF 2 -CF 2 -CF 2 binding (A) and (C) regions and some water and counter ions which are not in clusters. The hydrophilic region (C) consists of clusters containing most of the ionic exchange sites, counter ions and absorbed water. 23 In the cluster network model proposed by Gierke and Hsu, the structure is an inverted micelle in which the ion-exchange sites are separated from the fluorocarbon backbone thus forming spherical clusters, connected by short narrow channels. 24 When the membrane is dry, an average cluster has a radius of about 1.8 nm. In the swollen state the diameter increases to about 4 nm, each cluster being filled with about 1000 water molecules and the connecting channels have a diameter and a length of about 1 nm, as shown in Figure 3.4. Figure Gierke s cluster network model of Nafion membranes. Thus, with increasing water content the clusters grow and form transitory interconnections with each other. This network of collapsed channels leads to a percolation-type phenomenon. Gierke and Hsu also used the percolation theory to correlate the electrical conductivity with the water content of the membrane, expressed as λ, i.e. the number of water molecules per sulfonic group. According to this theory, there is a critical amount of water available in the membrane, below which ion transport is extremely difficult due to the absence of extended pathways. 25 The percolation threshold in Nafion is around λ = 2, as shown from conductivity data

40 3-Proton Exchange Membranes Figure 3.5 shows a schematic representation of the hydration states for Nafion membrane. At low hydration level, i.e. λ in the range 1-2, it is reasonable to consider that all water molecules absorbed by Nafion membrane are associated with the sulfonate heads because of the hydrophobic nature of the backbone and the hydrophilic nature of the sulfonic groups. Moreover, the hydronium ions will be localized on the sulfonate heads and the conductivity will be extremely low being the amount of water absorbed insufficient for the formation of a continuous water phase. 26 For λ in the range 3-5, the counterion clusters continue to grow and, as λ approaches 5, the membrane becomes more conductive because some counterion clusters may connect, but there is still insufficient water for all clusters to coalesce. Molecular dynamics simulations indicate that 5 water molecules form the primary hydration shell for the sulfonic groups and any additional water molecules are not as strongly bound and thus form a free phase. 27 For λ 6, counterion clusters coalesce to form larger clusters and eventually a continuous phase is formed, and the conductivity threshold is overcome. 34

41 3-Proton Exchange Membranes λ = 1 λ = 2 λ = 3 λ = 4 λ = 5 λ = 6 λ = 14 Figure Schematic hydration diagram for Nafion. Some variations on this hydration scheme are expected for the sulfonated polyetherketones membranes. Among other factors, the number of water molecules in the primary hydration shell will vary according to the strength of the sulfonic acid groups, and the distance between sulfonate heads will affect the conductivity threshold, which will vary with the amount of water needed to connect the clusters. Sulfonated polyetherketones have less pronounced hydrophobic/hydrophilic separation with respect to Nafion because their backbone are less hydrophobic and flexible and their sulfonic acid groups are less acidic and therefore, also less polar. As a consequence, in sulfonated polyetherketones microstructure narrower channels and a less-connected network of clusters are present, resulting in a higher dependence of the transport properties on water content due to percolation concepts

42 3-Proton Exchange Membranes A schematic representation of the microstructure of sulfonated polyetherketone compared with that of Nafion is reported in Figure 3.6. Figure Schematic representation of the microstructures of Nafion and a sulfonated polyetherketone. Both Nafion and sulfonated polyetherketones membranes exhibit Schroeder s paradox, an observed difference in the amount of water absorbed by a liquidequilibrated membrane and a saturated vapour equilibrated membrane, with both reservoirs at the same temperature and pressure. 29 Schroeder s paradox can be explained by models showing the structural changes of Nafion membrane as a function of water content, as shown in Figure 3.7. For saturated vapour water-equilibrated membrane (λ = 14), the water vapour does not have enough energy to condense in the channels due to their hydrophobicity. 36

43 3-Proton Exchange Membranes Conversely, the water uptake increases for the liquid water-equilibrated membrane (λ = 22) because the liquid water has enough pressure and energy to infiltrate and expand the channels. Thus, the reason for the difference in water uptake is due to the filling and expansion of the channels connecting the clusters. It is essentially a structural, not an equilibrium, issue. 30 λ = 0 λ < 2 λ = 14 λ = 22 Figure Evolution of the membrane structure as a function of water content. The gray area is the fluorocarbon matrix, the black is the polymer side-chain, the light gray is the liquid water, and the dotted line is a collapsed channel. 37

44 3-Proton Exchange Membranes 3.4 Proton Transport Proton Conduction Mechanisms Proton transport proceeds through the membrane following two main mechanisms. 31 The first is the vehicle mechanism where the proton diffuses together with the vehicle water. The counter diffusion of unprotonated water allows the net transport of protons. 32 Therefore, the observed conductivity depends on the rate of vehicle diffusion and it can be expressed as a function of water self-diffusion coefficient (D H2O ) that represents a measure of the average mobility of water in the membrane. The other principal mechanism is known as Grotthuss mechanism or proton hopping or structure diffusion. In this process, the water molecules show pronounced local dynamics but reside on their sites. The process consists of two steps: (1) proton transfer from one water molecule to the other by hydrogen bonds; (2) consequent reorientation of water dipoles that results in the formation of an uninterrupted trajectory for proton migration. The schematic description of two typical proton conduction mechanisms is shown in Figure Figure 3.8 Simplified scheme of the proton transfer in Nafion by the Grotthuss mechanism (solid lines) and the vehicle mechanism (dotted lines). 38

45 3-Proton Exchange Membranes The prevalence of one or the other mechanism depends on the hydration level of the membrane. 8 At low water content, (in Nafion λ = 2-3), the relative population of hydrogen bonds is low as well and the conductivity follows the vehicle mechanism. 39

46 3-Proton Exchange Membranes Proton Conductivity and Water Diffusion equation: Proton conductivity (σ) is related to proton mobility (D σ ) by the Nernst Einstein 2 cf σ = D σ (Eq. 3.1) RT where c is the proton concentration; F is Faraday s constant; R is the universal gas constant and T is the thermodynamic temperature. 34 Moreover, proton conductivity is strongly dependent on the membrane water content because proton displacement occurs through water-assisted mechanisms, hence the water self-diffusion coefficient (DH 2O) must be also considered. At high water content, the proton mobility (D σ ) is higher than D H2O and this difference increases with increasing water content, for both Nafion and sulfonated polyetherketones. The gap between the two coefficients indicates that some intermolecular proton transfer is involved in the mobility of protonic charge carriers via Grotthuss mechanism. 35 At very low hydration levels, the decrement of dissociation degree of the sulfonic acid groups leads to a stronger decrease of D σ compared to D H2O with decreasing of water content. This effect is more pronounced in sulfonated polyetherketones as a consequence of their less acidic sulfonic groups with respect to Nafion. Hence, the proton conductivity, that is related to D σ (Eq. 3.1), decreases much more severely with decreasing hydration level in sulfonated aromatic polymers than in Nafion. 27 Another phenomenon linked to membrane conductivity is electro-osmotic drag, the process whereby a certain number of water molecules associated with a proton are dragged through the liquid phase of the membrane. The electro-osmotic drag coefficient (K drag ) represents the number of water molecules dragged per proton and depends on the type of complex that water forms with a proton. At low degrees of hydration, the only complex that can form and be transported is the hydronium ion (H 3 O + ), that corresponds to an electro-osmotic drag coefficient of Thus, the conductivity and the electro-osmotic flow are due to H 3 O + that displaces 40

47 3-Proton Exchange Membranes from acid site to acid site through the clusters and across the collapsed channels, as schematically shown in Figure Figure Scheme of the transport mechanism at low water content. When the water content increases, the channels form continuous water pathways from cluster to cluster and from one side of the membrane to the other. The K drag increases with increasing hydration level because larger structures, such as Zundel (H 5 O + 2 ) and Eigen (H 9 O + 4 ) ions, can form. 37 These ions diffuse through the well interconnected liquid phase of membrane, and more water molecules are dragged per proton. The transport occurs by continuous formation of hydrogen bonds between the proton, its environments and water molecules, as schematically shown in Figure Figure Scheme of the transport mechanism at high water content. For a given water content, sulfonated polyetherketones show a lower electroosmotic drag coefficient than Nafion, due to their narrower water channels. 38 As important consequence, their lower K drag facilitate the water management under cell operating conditions and their lower water and methanol crossover is advantageous with respect to that of Nafion, for DMFC applications. 39,40,41 41

48 3-Proton Exchange Membranes Activation Energy Proton conduction in a polymer membrane is a thermally activated process. At temperatures below and close to the glass transition temperature (T g ) of the polymer, the conductivity (σ) obeys an Arrhenius-type law: σ E a = A e RT (Eq. 3.2) where A is a constant proportional to the number of charge carriers; E a is the activation energy; R is the universal gas constant; T is the absolute temperature. 42 Activation energy represents the minimum energy required for proton transport through the membrane and it can be determined experimentally via Arrhenius plots for proton conductivity. Activation energy value depends on membrane water content, occurring the proton displacement through water-assisted mechanisms. For Nafion in a dry state, that is containing only residual water molecules, E a falls in the range ev. In the highly swollen membrane, the activation energy is 0.1 ev, which is close to the bulk water value. At intermediate degrees of wetting, the activation energy varies between these two values

49 3-Proton Exchange Membranes 3.5 Synthetic Approaches The synthesis of functionalized polymers represents a very promising way to obtain proton-conducting membranes having the desired features for fuel cell applications because it allows to prepare tailored materials for specific applications. 44,45,46 Several methods have been developed to prepare proton-conducting electrolytes, including total synthesis from monomer building blocks, direct sulfonation of a polymer backbone, and grafting of functional groups onto a polymer main chain. A schematic diagram is shown in Figure Figure Development of proton-conducting membranes by the modification of thermostable polymers. Good tailoring of polymer composition is possible by the direct synthesis of a new polymer from monomer building blocks. 48,49 Control of the position, number, and distribution of protogenic functions along the polymer backbone could allow tuning of the microstructure and correlated properties such as conductivity and degree of swelling. However, the range of polymers that can be produced is limited by the commercial availability or the ease of preparation of the corresponding monomers. 43

50 3-Proton Exchange Membranes Moreover, multistage syntheses could increase costs rendering the polymer less attractive for industrial use than one prepared by direct sulfonation. Consequently, there is a great interest in using arylene main-chain polymers, such as poly(ether ketone)s (PEK), polysulfones (PSU) and polyimides (PI), because they are high-performance and low-cost materials that can be sulfonated by direct reaction with an appropriate reagent. In the last years, a large number of these polymers have been sulfonated to obtain proton-conducting electrolytes. 50,51,52 Some of these sulfonated polymers are shown in Figure The most attractive site for sulfonation depends both on the polymer structure and the directing effect of substituent groups, and advantages of simplicity and reproducibility of the sulfonation reaction are associated with this inherent reactivity. Figure a) Sulfonated polysulfone; b) Sulfonated polyetheretherketone. A general problem with use of the sulfonated polymers, having the sulfonic groups bound directly to the aromatic chain, is the undesirable swelling of membrane when a certain degree of sulfonation is exceeded. Several strategies have been used to overcome these problems, including the separation of polymer main chain and sulfonic acid groups by placing the protogenic groups on short pendant side chains, 53,54 and the synthesis of aromatic polymer chains cross-linked covalently by organic spacers. 55 An emerging strategy is the development of hybrid materials in which organic and inorganic components are combined at molecular level. The nanostructure, the degree of organization and the properties that can be obtained for such materials certainly depend on the chemical nature of their components, but they also rely on the synergy between them. 44

51 3-Proton Exchange Membranes The nature of the interface between the organic and inorganic components has been used to classify the hybrids into two different classes: 56 Class I corresponds to systems where there are no covalent bonds between the organic and inorganic components, but only exchange interactions such as van der Waals forces, hydrogen bonding or electrostatic forces. These hybrids are also called composites and will be discussed in detail in Section 3.6. Class II corresponds to systems where the organic and inorganic components are linked through covalent bonds. These hybrids are called nanocomposites and can be composed of structural units, which span the whole range from inorganic glass type to compositions with a high content of organic network similar to carbon based polymers, as can be seen in Figure Figure Structural units of Class II hybrids. Two types of Class II hybrids can be further distinguished: 58 Organic-Inorganic Polymers (OIPs) that have carbon atoms in the main chain/network and inorganic groups linked to the organic structures. Inorganic-Organic Polymers (IOPs) that have inorganic elements in the main chain/network and organic side groups. 45

52 3-Proton Exchange Membranes The organic-inorganic polymers open very broad and promising horizons to obtain proton-conducting electrolytes with the desired features. The functionalization of organic polymers, where inorganic moieties are grafted onto backbone by strong covalent bonds, represents an interesting synthetic methodology to tailor molecular structures controlling the dosage and dispersion of inorganic clusters. Among the possible functionalizations, the introduction of silicon-containing moieties appears to be extremely interesting to improve water management in PEMFCs operating at high temperatures. In these structures additional inorganic crosslinking, due to Si-O-Si bonds, can be achieved by sol-gel process. 59 Figure 3.14 shows an OIP, where aromatic chains of poly(ether ether sulfone)-co-(ether sulfone) (PEES-PES) are cross-linked via SiO 4 bridges. Figure Depiction of PEES-PES chains that are cross-linked through the end groups via SiO 4 bridges. Over the last decade, numerous studies were also performed for designing inorganic-organic polymers for different applications. 60,61,62 When the organic groups of IOPs are reactive, additional organic crosslinking/polymerization is possible. If classical sol-gel processing of modified metal alkoxides is followed by organic polymerization or crosslinking, the resulting hybrid materials are called organically modified ceramics (ORMOCERs). 46

53 3-Proton Exchange Membranes An interesting type of these materials are organic-modified silicates (ORMOSILs) that show good thermal and mechanical stability due to the high Si-O bond strength, making them suitable for PEMFC applications. 63,64 An ORMOSIL with siloxane backbone and aryl-containing side chains is shown in Figure Figure Organically Modified Silicate. The basic idea of both OIPs and IOPs is to obtain a synergic effect between inorganic and organic phases, not achievable by physical mixing of organic and inorganic materials, as usually made in classical composites (Class I hybrids). The building principle of these polymers is based on different types of precursor molecules that can be classified by their network forming or modifying. This field of materials research mainly arises from chemists skills and demonstrates the major role played by chemistry in the preparation of membrane materials for fuel cell applications. 47

54 3-Proton Exchange Membranes 3.6 Composite and Blend Approaches An effective way to achieve low-humidity and high-temperature operating electrolytes is to prepare composite membranes (Class I hybrids) by casting the polymer solution with an inorganic component. The solid inorganic compounds include hygroscopic oxides, e.g. silica (SiO 2 ) and titania (TiO 2 ), and inorganic proton conductors, such as zirconium phosphates and heteropolyacids. 4 The growth of inorganic particles in situ in a polymer solution provide a means of modifying the polymer microstructure because the membrane structure reforms at the same time as the precipitation of the inorganic and/or organic components. Several approaches have been used to generate organic-inorganic composite membranes, such as: (1) casting a bulk mixture of powder or colloidal inorganic components with a polymer solution, or (2) in situ formation of inorganic components in a polymer solution. To prepare uniform and nonporous membranes, size and dispersion of the solid particles are of special importance. For bulk mixing, the use of colloidal suspension of inorganic component enable the nanosized particles to be dispersed in the formed membranes. 65 On the other hand, the formation of inorganic components in situ in a polymer matrix can provide many advantages, such as a homogeneous dispersion and a careful morphological control. This second approach was extensively studied by Mauritz et al. that used a hydrolytic sol-gel route to generate the inorganic network inside the polymer phase. Core of the preparation process using sol-gel techniques is the hydrolysis of an element or metalorganic precursor compound in the matrix of the polymer, yielding an inorganic oxide or oxide/hydroxide network in the polymer phase. 66,67 Composite membranes containing hygroscopic oxide particles, such as SiO 2, 68,69 TiO 2, 70,71 SnO 2, 72 and ZrO 2, 73 have been investigated by several research groups. Inclusion of these compounds in polymer matrix lead to increase of membrane hydration, due to the water adsorption on the oxide surface, and therefore electrolytes with reduced conduction losses at low-humidity operating conditions can be obtained. Moreover, composites membranes containing inorganic proton conductors were widely studied. 74,75 Inclusion of these bifunctional particles, that are both proton 48

55 3-Proton Exchange Membranes conducting and hydrophilic compounds, in the polymer matrix creates additional pathways for the proton conduction and provides hydrogen bonding sites for water in the membrane. 76 Among the solid proton conductors, zirconium phosphates are of special interest for developing high temperature composite membrane because they exhibit good proton conductivity in a temperature range up to 300 C. These compounds have been incorporated into polymeric membranes, obtaining electrolytes with good mechanical 77, 78 strength and stability of proton conductivity. Therefore, several advantages can be obtained by using composite membranes, such as: i) improving the self-humidification of the membrane at the anode side; ii) suppressing the fuel crossover, e.g. methanol in DMFC; iii) improving the mechanical strength of membranes without excessively sacrificing proton conductivity. 79 The main disadvantage of such composite systems is related to the fact that it is very hard to obtain homogenous systems, where the inorganic particles are well dispersed in the polymeric matrix. Polymer blending represents another versatile and inexpensive way to obtain proton exchange membranes with certain targeted properties for fuel cell application. Blend membranes are prepared by mixing together two or more organic polymers with different microstructural, thermal, mechanical and conductivity properties to obtain a synergic effect among different components. Ionomer membrane types have been developed which show the different types of interaction forces between the blend components, such as van der Waals interactions, hydrogen bridge, and dipole-dipole interactions, as schematically shown in Figure

56 3-Proton Exchange Membranes Figure Specific interaction between macromolecules. Polymer blends, in which van der Waals/dipole-dipole interactions occured between two components, have been prepared starting from a variety of ion-exchange materials, such as sulfonated polysulfone (SPSF), sulfonated polyphenyleneoxide (SPPO) and sulfonated polyetheretherketone (SPEEK) with other materials, such as polyethersulfone (PES) and polysulfone (PSF). 81,82 The basic idea for this membrane type was to reinforce the sulfonated polymer membrane with unmodified polymer, due to the fact that polymers lose much of their mechanical strength by sulfonation. Blend membranes with increased bond strength between two components have been prepared by mixing silylated and sulfonated polyphenylsulfone (SiSPPSU) and SPEEK. Due to the similar aromatic structure of two polymer, π π interactions occurred between blend components, together with electrostatic interactions connected with hydrogen bridge and dipole-dipole interactions. These electrolytes have shown good proton conductivity and mechanical strength. 83,84 Other types of blend membranes, with hydrogen bridge interactions between two polymers, have been prepared starting from polyetherimide (PEI) and SPEEK or sulfonated polyetherketoneketone (SPEKK), obtaining electrolytes with good proton conductivity. 85,86 An interesting class of proton-conducting membranes is made up of ionically cross-linked acid-base blends. The interaction forces between the acidic and the basic 50

57 3-Proton Exchange Membranes components include electrostatic and hydrogen bridge interactions. The structure of acid-base blend membranes is depicted schematically in Figure Figure Scheme of Ionically cross-linked acid-base blend membranes. Several types of ionically cross-linked acid-base blends have been investigated. 87,88 The used acidic polymers are sulfonated polysulfone (SPSF), sulfonated polyethersulfone (SPES), or sulfonated polyetheretherketone (SPEEK). Some basic polymers are commercially available, such as PBI, polyethyleneimine (PEI), and poly(4-vinylpyridine) (P4-VP). New basic polymers have been synthesized by modifying the PSF backbones with amino (-NH 2 ) or dimethylamino (-N(CH 3 ) 2 ) groups. 89 Combinations of SPEEK/PBI, SPEEK/P4VP, SPEEK/PSF(NH 2 ) 2, SPSF/PBI, and 90, 91 SPSF/P4VP have been also explored. 3.7 References 1) J. Larminie, A. Dicks, Fuel Cell Systems Explained, John Wiley & Sons Ltd, Ontario, ) M. Doyle, G. Rajendran, Handbook of Fuel Cells, vol. 3, W. Vielstich, A. Lamm, H.A. Gasteiger ( Eds.), Wiley, Chichester, England, ) M.A. Hickner, B.S. Pivovar, Fuel Cells 2005, 5, ) L. Qingfeng, H. Ronghuan, J.O. Jensen, N.J. Bjerrum, Chem. Mater. 2003, 15, ) L.P. Hammett, Physical Organic Chemistry, 2 nd Ed., McGraw-Hill Ltd, Kogakusha, Tokyo,

58 3-Proton Exchange Membranes 6) M. Hickner, H. Ghassemi, Y.S. Kim, B.R. Einsla, J.E. McGrath, Chem. Rev. 2004, 104, ) G. Alberti, R. Narducci, M. Sganappa, J. Power Sources 2008, 178, ) K.D. Kreuer, S. Paddison, E. Spohr, M. Schuster, Chem. Rev. 2004, 104, ) A.S. Aricò, S. Srinivasan, V. Antonucci, Fuel Cells 2001, 1, ) W. G. Grot, US Pat., , ) B. Smitha, S. Sridhar, A.A. Khan, J. Membr. Sci. 2005, 259, ) O. Savadogo, J. Power Sources 2004, 127, ) R.W. Dyson, Ed. Specialty Polymers, Blackie Academic & Professional London, ) P. Xing, G.P. Robertson, M.D. Guiver, S.D. Mikhailenko, K. Wang, S. Kaliaguine, J. Membr. Sci. 2004, 229, ) T. Fu, Z. Cui, S. Zhong, Y. Shi, C. Zhao, G. Zhang, K. Shao, H. Na, W. Xing, J. Power Sources 2008, 185, ) J. Kerres, Fuel Cells 2006, 3, ) B. Lafitte, L.E. Karlsson, P. Jannasch, Macromol. Rapid Commun. 2002, 23, ) C.S. Karthikeyan, S.P. Nunes, K. Schulte, Adv. Engineering Mater. 2006, 8, ) P. Mustarelli, E. Quartarone, S. Grandi, A. Carollo, A. Magistris, Adv. Mater. 2008, 20, ) J. Divisek, M. Eikerling, V. Mazin, H. Schmitz, U. Stimming, Y.M. Volfkovich, J. Electrochem. Soc. 1998, 145, ) R.S. McLean, M. Doyle, B.B. Sauer, Macromolecules 2000, 33, ) K.A. Mauritz, C.J. Hora, A.J. Hopfinger, Ions in Polymers; Ed. A. Eisenberg, ACS Advances in Chemistry Ser. No. 187, American Chemical Society: Washington, DC, ) H.L Yeager, A. Steck,. J. Electrochem. Soc. 1981, 128, ) W.Y. Hsu, T.D. Gierket, Macromolecules 1982, 15, ) W.Y Hsu, T.D. Gierke, J. Membr. Sci. 1983, 13, ) J. Fimrite, H. Struchtrup, N. Djilali, J. Electrochemical Soc. 2005, 152, A ) J. Elliot, S. Hanna, A.M.S. Elliot, G.E. Cooley, Phys. Chem. Chem. Phys. 1999, 1, ) K.D. Kreuer, J. Membr. Sci. 2001, 185,

59 3-Proton Exchange Membranes 29) T.A. Zawodzinski, T.E. Springer, J. Davey, R. Jestel, C. Lopez, J. Valerio, S. Gottesfeld, J. Electrochem. Soc. 1993, 140, ) A.Z. Weber, J. Newman, J. Electrochem. Soc. 2003, 150, A ) K.D. Kreuer, Chem. Mater. 1996, 8, ) K.D. Kreuer, W. Weppner, A. Rabenau, Angew. Chem., Int. Ed. Engl. 1982, 21, ) P. Choi, N.H. Jalani, T.M. Thampan, R. Datta, J. Polym. Sci: Part B: Polym. Phys. 2006, 44, ) P.G. Bruce, Solid State Electrochemistry, Cambridge University Press, ) N. Agmon, Chem. Phys. Lett. 1995, 244, ) T.A. Zawodzinski, J. Davey, J. Valerio, S. Gottesfeld, Electrochim. Acta 1995, 40, ) K.D. Kreuer, Solid State Ionics 2000, 136, ) M. Ise, K.D. Kreuer, J. Maier, Solid State Ionics 1999, 125, ) M. Eikerling, Y.I. Kharkats, A.A. Kornyshev, Y.M. Volfkovich, J. Electrochem. Soc. 1998, 145, ) T. Okada, G. Xie, M. Meeg, Electrochim. Acta 1998, 43, ) X.M. Ren, W. Henderson, S. Gottesfeld, J. Electrochem. Soc. 1997, 144, L ) Y. Sone, P. Ekdunge, D. Simonsson, J. Electrochem. Soc. 1996, 143, ) M. Eikerling, A.A. Kornyshev, J. Electroanal. Chem. 2001, 502, 1. 44) B. Smitha, S. Sridhar, A.A. Khan, J. Membr. Sci. 2005, 259, ) M. Ulbricht, Polymer 2006, 47, ) Y. Yang, S. Holdcroft, Fuel Cells 2005, 5, ) J. Rozière, D.J. Jones, Annu. Rev. Mater. Res. 2003, 33, ) F. Wang, M. Hickner, Y.S. Kim, T.A. Zawodzinski, J.E. McGrath, J. Membr. Sci. 2002, 197, ) G. Maier, C-K Shin, G.G. Scherer. In 1 st Eur. PEFC Conference, ed. FN Büchi, GG Scherer, A Wokaun, Lucerne, Switerland: European Fuel Cell Forum, ) J.A. Kerres, J. Membr. Sci. 2001, 185, 3. 51) A. Dyck, D. Fritsch, S.P. Nunes, J. Appl. Polym. Sci. 2002, 86, ) S. Kaliaguine, S.D. Mikhailenko, K.P. Wang, P. Xing, G. Robertson, M. Guiver, Catalysis Today 2003, 82,

60 3-Proton Exchange Membranes 53) L.E. Karlsson, P. Jannasch, Electrochim. Acta 2005, 50, ) J.M. Bae, I. Honma, M. Murata, T. Yamamoto, M. Rikukawa, N. Ogata, Solid State Ionics 2002, 147, ) W. Zhang, V. Gogel, K. Friedrich, J. Kerres, J. Power Sources 2006, 155, 3. 56) C. Sanchez, F. Ribot, New J. Chem. 1994, 18, ) K.H. Haas, K. Rose, Rev. Adv. Mater. Sci. 2003, 5, ) H.L. Allcock, Adv. Mater. 1994, 6, ) C.H. Brinker, G. Scherer, Sol-gel Science, Academic Press, London, ) C. Sanchez, B. Lebeau, F. Chaput, J.P. Boilot, Adv. Mater. 2003, 15, ) R. Houbertz, L. Fröhlich, M. Popall, U. Streppel, P. Dannberg, A. Bräuer, J. Serbin, B. N. Chichkov, Adv. Engineering Mater. 2003, 5, ) L.C. Klein,Y. Daiko, M. Aparicio, F. Damay, Polymer 2005, 46, ) D. Gomes, I. Buder, S.P. Nunes, J. Polym. Sci: Part B: Polym. Phys. 2006, 44, ) M. Jeske, C. Soltmann, C. Ellenberg, M.Wilhelm, D. Koch, G. Grathwohl, Fuel Cells 2007, 1, ) G. Alberti, M. Casciola, U. Costantino, R. Vivani, Adv. Mater. 1996, 8, ) K.A. Mauritz, J.T. Payne, J. Membr. Sci. 2000, 168, ) S.P. Nunes, B. Ruffmann, E. Rikowski, S. Vetter, K. Richau, J. Membr. Sci. 2002, 203, ) B. Bonnet, D.J. Jones, J. Rozière, L. Tchicaya, G. Alberti, M. Casciola, L. Massinelli, B. Bauer, A. Peraio, E. Ramunni, J. New Mat. Electrochem. Systems 2000, 3, ) K.T. Adjemian, S.J. Lee, S. Srinivasan, J. Benziger, A.B. Bocarslya, J. Electrochem. Soc. 2002, 149, A ) M. Watanabe, H. Uchida, Y. Seki, M. Emori, J. Electrochem. Soc. 1996, 143, ) M.L. Di Vona, Z. Ahmed, S. Bellitto, A. Lenci, E. Traversa, S. Licoccia, J. Membr. Sci. 2007, 296, ) B. Mecheri, A. D Epifanio, E. Traversa, S. Licoccia, J. Power Sources 2008, 178, ) K.A. Mauritz, Mat. Sci. Eng. 1998, C6,

61 3-Proton Exchange Membranes 74) P. Staiti, A.S. Arico, V. Baglio, F. Lufrano, E. Passalacqua, V. Antonucci, Solid State Ionics 2001, 145, ) P. Costamagna, C. Yang, A.B. Bocarsly, S. Srinivasan, Electrochim. Acta 2002, 47, ) K.T. Adjemian, S. Srinivasan, J.Benziger, A.B. Bocarsly, J. Power Sources 2002, 109, ) G. Alberti, M. Casciola, D. Capitani, A. Donnadio, R. Narducci, M. Pica, M. Sganappa, Electrochim. Acta 2007, 52, ) L. Tchicaya-Bouckary, D.J. Jones, J. Rozière, Fuel Cells 2002, 2, ) G. Alberti, M. Casciola, Annu. Rev. Mater. Res. 2003, 33, ) J. A. Kerres, Fuel Cells 2005, 5, ) V. Deimede, G.A. Voyiatzis, J.K. Kallitsis, Q. Li, N.J. Bjerrum, Macromolecules 2000, 33, ) J. Kerres, A. Ulrich, F. Meier, T. Haring, Solid State Ionics 1999, 125, ) M. L. Di Vona, A. D Epifanio, D. Marani, M. Trombetta, E. Traversa, S. Licoccia, J. Membr. Sci. 2006, 279, ) E. Sgreccia, M. Khadhraoui, C. de Bonis, S. Licoccia, M. L. Di Vona, P. Knauth, J. Power Sources 2008, 178, ) S.D. Mikhailenko, S.M.J. Zaidi, S. Kaliaguine, J. Polym. Sci. Polym. Phys. 2000, 38, ) S. Swier, M.T. Shaw, R.A. Weiss, J. Membr. Sci. 2006, 270, ) J.A. Kerres, J. Membr. Sci. 2001, 185, 3. 88) J. Kerres, W. Zhang, L. Jörissen, V. Gogel, J. New Mat. Electrochem. Systems 2002, 5, ) M.D. Guiver, G.P. Robertson, S. Foley, Macromolecules 1995, 28, ) L. Jorissen, V. Gogel, J. Kerres, J. Garche, J. Power Sources 2002, 105, ) C. Hasiotis, V. Deimede, C. Kontoyannis, Electrochim. Acta 2001, 46,

62 4 Experimental 4.1 Materials and Methods This section presents a description of the starting materials and characterization methods of the polymers and membranes Starting Materials Polyetheretherketone (Victrex, PEEK, 450 P, MW = g/mol, 132 repeat units per mol), Polyphenylsulfone (Solvay, PPSU-RADEL R, 5100 P, MW = g/mol, 115 repeat units per mol) and all other chemicals (Aldrich) were used asreceived. Anhydrous tetrahydrofuran (THF) and dichloromethane (CH 2 Cl 2 ) were prepared according to literature procedures. 1 PEEK Polyetheretherketone is a semicrystalline thermoplastic with the glass transition temperature (T g ) of 150 C. It shows extraordinary mechanical properties and exhibits good chemical resistance in many environments, including alkalis, aromatic hydrocarbons and alcohols. PEEK consists of alternating hydroquinone and diphenyl carbonyl segments, as shown in Figure 4.1. Figure Monomeric repeat unit of PEEK. 56

63 4-Experimental Aromatic electrophilic substitution may occur easily on the hydroquinone unit due to the presence of activating and ortho-directing ether groups, in contrast with the diphenyl carbonyl portion which has a low electron density, due to the electronwithdrawing effect of carbonyl group. PPSU Polyphenylsulfone is an amorphous polymer with a glass transition temperature (T g ) of 220 C. It shows good oxidative resistance and is among the most heat resistant of current commercial thermoplastics. 2 PPSU consists of alternating bisphenol and diphenyl sulfone segments, as shown in Figure 4.2. Figure Monomeric repeat unit of PPSU. The strong electron-withdrawing effect of the sulfone units gives an acidic character to the aromatic hydrogens in the ortho position. Thus, the carbons at these sites may be metalized by the use of a strong base. On the other hand, the electron-donating effect of the ether linkages activates the phenylene rings of the bisphenol part towards electrophilic substitution. 3 Thus, for both PEEK and PPSU several functionalizations are possible since their chemical structure offers the possibility of electrophilic as well as nucleophilic substitution, increasing their application areas. In particular, both polymers show proton conductivity when sulfonated, thus being potential electrolytes for PEMFCs. 4 57

64 4-Experimental Methods The chemical structure of synthesized polymers was investigated by nuclear magnetic resonance (NMR) and attenuated total reflectance Fourier transform infrared (ATR/FTIR) spectroscopies, and their thermal stability was evaluated by thermogravimetric analysis (TGA). Membranes features were investigated by water uptake (W.U.) measurements, electrochemical impedance spectroscopy (EIS) and tests in a prototype fuel cell. Nuclear Magnetic Resonance Spectroscopy 1 H and 13 C NMR spectra were recorded with a Bruker Avance 300 spectrometer operating at 300 and 75.4 MHz, respectively, using DMSO-d6 as solvent. Chemical shifts (ppm) were referenced to tetramethylsilane (TMS). ATR/FTIR Spectroscopy Infrared spectra were collected in the range cm 1, on a Nicolet 8700 E.S.P. with a Golden Gate MK2 Diamond Specac cell. Spectra were recorded positioning the samples on cell platform operating at room temperature (32 scans, 2 cm 1 resolution). Thermogravimetric Analysis Thermogravimetric analyses were carried out using a Netzsch STA 409 thermobalance, in air flow (80 ml/min) with a heating rate of 2 C/min in the range C, using platinum sample holders. Differential Scanning Calorimetry The measurements were carried out by using DSC 7 Perkin-Elmer Differential Scanning Calorimeter. Thermograms were recorded under N 2 flow (20 ml/min), placing the samples in aluminum pans. The samples were preheated from 10 C to 160 C at 20 C/min to remove moisture, then cooled to 90 C, and reheated from that temperature to 250 C at 20 C/min. 58

65 4-Experimental Membranes preparation Literature conductivity values of proton exchange membranes with similar composition show large discrepancies suggesting that the preparation procedure has a significant influence on the electrical properties of these electrolytes. Membranes made up of sulfonated PEEK and/or PPSU are usually obtained by casting the polymer solution on a flat surface. The solvents commonly used to prepare the solution are: N,N-dimethylacetamide (DMAc), dimethylformamide (DMF), dimethyl sulfoxide (DMSO) and N-methyl-2-pyrrolidone (NMP). 5 However, it must be considered that NMP and DMSO are solvents with higher boiling temperature (202 C and 189 C, respectively) compared with DMF (153 C) and DMAc (166 C). As a consequence, it is particularly difficult to dry the NMP or DMSO cast membranes. On the other hand, amide-based solvents, i.e. DMAc and DMF, can form strong hydrogen-bonds with the sulfonic acid groups of the polymer and these interactions may significantly decrease the number and/or mobility of protons, reducing membrane conductivity. In this work, DMAc was chosen as solvent because it has relatively low boiling point and forms hydrogen-bonds with sulfonic acid groups only above 140 C. 6 In a typical experiment for the preparation of the membranes of sulfonated PEEK (S 0.5 PEEK) or hybrid polyphenylsulfone derivative (HSiSPPSU), 400 mg of polymer was dissolved in DMAc (20 ml). For the preparation of the blend membranes, 400 mg of S 0.5 PEEK was dissolved in DMAc (20 ml) and the appropriate weight of silylated and sulfonated PEEK (PhSi 0.1 S 0.9 PEEK) or PPSU (PhSi 0.2 S 2 PPSU) was dissolved in the same solvent (10 ml), then the two solutions were mixed. In all cases, the resulting mixture was heated under stirring until the volume was reduced to 5 ml, then cast onto a Petri plate (5.5 cm of diameter) and heated to 80 C for 24 hours. The thickness of the membranes was in the range µm. The membranes were activated by reflux in 0.5 M H 2 SO 4 for 1 hour and in distilled water for 1 hour. Before measurements, the samples were equilibrated at RT, at ambient humidity (~ 50%) for 48h. 59

66 4-Experimental The activation of the membranes in sulphuric acid was needed to remove the protonated casting solvent, and the subsequent treatment with boiling water was carried out to remove residual sulphuric acid and to promote the water uptake of membranes. In fact, it is well known that an adequate water uptake is required for good proton conductivity, since proton transport occurs through a water-assisted mechanism. Solvent uptake Water and methanol uptake measurements were performed on the membranes to evaluate their hydration behavior and methanol affinity. Disks were cut from the membranes, dried overnight at 80 C and immediately weighted to obtain the dry weights (W dry ). Two different procedures were then carried out to measure the water uptake, at 100% relative humidity (RH). In the first one, dry disks were immersed in distilled water, at room temperature, for 24 hours. The samples were then rapidly dried with adsorbent paper and weighted. In the second procedure, dry samples were exposed at saturated water vapour pressure at room temperature, overnight, and then weighted. Water uptake values, expressed in percentage units, (W.U.%) were calculated using the following equation: 7 Wwet WU..% = W dry (Eq. 4.1) where W dry and W wet are the dry and wet weights, respectively. Methanol uptake measurements were also carried out. Dry disks were immersed in absolute methanol, at room temperature, for 24h. The samples were then rapidly dried with adsorbent paper and weighted to obtain the wet weights (W wet ). 60

67 4-Experimental Conductivity Measurements Proton conductivity of the membranes was measured by electrochemical impedance spectroscopy (EIS) using a Multichannel Potentiostat VMP3 (Princeton Applied Research). An applied voltage of 20 mv and a frequency range of 500 KHz to 10 Hz were used. Membranes were sandwiched between commercial electrodes (E-Tek ELAT HT 140E-W with a platinum loading of 5 g m -2 ), as shown in Figure 4.3. A teflon ring was used as spacer to avoid the short circuit between the electrodes. Porous steel Teflon Membrane Porous steel ELAT Figure Electrodes and membrane assembly for EIS measurements. Proton conductivity of the membranes was measured as a function of temperature, at 100% relative humidity (RH). Before measurements, the samples were kept overnight at room temperature (RT) and at 100% RH, and conductivity values were then obtained in the range C, at saturated water vapour pressure. Proton conductivity of the membranes was also measured as a function of RH, at a fixed temperature. Controlled RH environments were obtained using the homemade apparatus that is schematically shown in Figure

68 4-Experimental Figure Scheme of the double chamber apparatus. It is constituted of two compartments one for the lodging of the test cell and the other containing water. The relative humidity was controlled by the different temperature between the two compartments, T 1 and T 2, where T 1 T 2, because water vapour pressure depends on temperature and RH% is expressed by the following equation: RH = P ( T) (Eq. 4.2) o HO 2 1 % 100 o P HO( T 2 2) where P H 2O (T 1 ) is the saturated water vapour pressure at the temperature of the watercontaining compartment, while P H 2O (T 2 ) is the saturated water vapour pressure at the temperature of the cell-containing compartment. Electrical heaters were used to control the two temperatures, where T 2 was kept constant, whereas T 1 was varied to achieve different RH% values that were monitored by a humidity sensor (Vaisala HMP237) in the % range. The resistance of the membranes was calculated by a linear fit of the Nyquist plots in their linear portion. Conductivity values (σ) were obtained from the resistance values (R), the thickness of the membrane (l) and the electrode area (A), using the following equation: σ = l R A (Eq. 4.3) 62

69 4-Experimental Direct Methanol Fuel Cell Testing Membrane electrode assemblies (MEAs) were fabricated using commercial electrodes E-TEK, A-6 ELAT/SS/PtRu PTFE/0.8N, custom 3 mg cm -2 loading using 60% HP Pt:Ru Alloy (1:1) on Vulcan, 30 wt % PTFE with 0.8 mg cm -2 Nafion (anode), and A-6 ELAT/SS/Pt PTFE/0.8N custom 2 mg cm -2 loading using 40% HP Pt on Vulcan, 30 wt % PTFE with 0.8 mg cm -2 Nafion (cathode). MEAs were prepared by hot pressing using an SPECAC hot press. Membrane and anodic electrode were placed in a teflon sheet and pressed for 1 min at 80 C under a load of 0.62 MPa. After that, the cathode was put on the other side of the membrane. Proper interfacial adhesion in the electrode/membrane/electrode was achieved pressing the sandwich at 80 C under a load of 0.62 MPa for 5 min. Figure 4.5 shows a picture of one of the prepared MEAs. Figure A typical MEA. Fuel cell experiments were carried out in a 5 cm 2 single cell (ElectroChem) using a homemade fuel cell test station. The test station was equipped for the independent temperature control of the reactant gas (O 2 ), methanol solution and single cell. Flow rate of the gas was regulated using mass-flow controller (MKS PR4000, 300 sccm), the total pressure of the gas was controlled using a back-pressure regulator. Flow rate of methanol solution (2 M) was monitored by a KNF's Model FEM 1.08 Liquid Metering Pump with control board (1 ml min -1 ). All experiments were performed fixing at 0.2 MPa the total pressure in the cell and at 100 C the cell temperature. Polarization data were recorded using a Multichannel Potentiostat VMP3 (Princeton Applied Research). 63